[Oriya-group] Re: [Orissa-IT] Re: Conversion from OR_TTSarala to Unicode

[Vivekananda Pani]

On Thu, 23 Dec 2004 23:16:50 -0800 (PST), Hariram Pansari
<hrpansari at yahoo.com> wrote:
> 
> --- gora_mohanty <gora_mohanty at yahoo.co.in> wrote:
> 
> > Pansaribabu, if you really want to help out, for any
> > given font
> > (Devanagari, Oriya, or any language, but please do
> > not redo work on
> > OR_TT*, OR1_TT*, ORB_TT*) please provide me with
> > maps that convert
> > each glyph to a sequence of Unicode characters.
> > 
> I will upload (to the files section of this group) a
> draft/beta convertion table from ORBW-TTSarala to
> ISCII and vice-versa in Complete_Syllable to
> Complete_Syllable format.

A query here. When you make conversion tables, do you laboriously type
each possible syllable and then type its ISFOC quivalent for OR, ORB,
ORW and ORBW? Well, may i suggest to extend a small help here. It will
be a matter of minutes to write a small command line C program that
can generate "all" possible ISCII syllables of 2 or 3 or 4 consonant
combinations and matra and modifiers (whatever we need, we can put in
a rule file). You can then use the ISM conversion tools to convert
these ISCII to their respective ISFOC (assuming that all the forward
ISCII to ISFOC for various oriya font layouts work well). Merge these
with space or tab as delimiters. The table is ready. I can help in
such programs to create syllables. Or, better still, you can send me
your syllable requirement and I will give back the syllable list in
ISCII (thousands or hundreds of thousands as the need/cases may be).
Please let me know if this may be helpful.

> With experience of 12 years, since 1992, I strongly
> say that latest ORBW-TTMukta Font of CDAC is the best,
> all glyphs are made clear and Web compatible. The old
> version of ORB-TTMukta-1998 (of DLL-16) is being used
> only by Srujanika. All other Oriya user's have
> upgraded (to DLL-32). So I do not see any wider
> usefulness of our effort/time to be invested for it.
> 
> Directly from 8bit-font-glyph-sets to Unicode is not
> possible, because presently most of users has to stick
> to Windows98 OS (as high cost involves for upgrade),
> which does not support 16Bit encodings. And most of
> the users not familiar with linux.
> 
> Better you may work presently for a converter from
> OR_ISCII to Unicode. (The ISCII-Unicode
> converter(proprietory) provided by CDAC has lot of
> bugs.) Most of the Oriya Softwares like Akruti,
> Indica, Modular... others.... now also provides "SAVE
> AS ISCII" or "Export to ISCII" feature. This will be
> more useful for them also.

> OR_ISCII to Unicode converter, I think will be very
> easy to design. I Think this will take an hour or two
> for you.

You are right. I can provide the tool. Can you tell me if it needs to
be like a command line program accepting two file names (Input and
output)?

> I am trying to upload a draft/beta conversion table
> from OR_ISCII to Unicode.
This can reduce the programmer's job enormously..:)

regards,
Vivek.